Skip to main content

New GPU-Z Update Detects Enigmatic Tiger Lake Server CPUs, GeForce LHR GPUs

Intel
(Image credit: Intel)

At times a new release of hardware detection software not only improves functionality and detection capabilities, but also reveals unknown or upcoming hardware. This has happened again with the latest GPU-Z utility that adds detection for added a mysterious 'Tiger Lake Server' processor along with Intel's upcoming Alder Lake CPUs, Nvidia's GeForce LHR GPUs (with correct hash rates), and the GeForce RTX 3060 based on the GA104 chip.  

We have never heard of Tiger Lake Server CPU before. Still, keeping in mind that the processor would presumably be made using Intel's 10nm SuperFin fabrication process that was architected primarily for client PCs, we don't really expect Intel to design brand-new server-grade silicon based on the Tiger Lake design (e.g., Willow Cove microarchitecture). 

This doesn't mean Intel couldn't merely use its existing quad-core Tiger Lake (UP3, UP4, H35) or eight-core Tiger Lake-H designs to address servers. For example, earlier this year, Intel quietly launched its Tiger Lake B-series processors with a 65W TDP and Thermal Velocity Boost (TVB) for high-performance small form-factor desktops. 

The server market is very diverse nowadays, so several server designs could benefit from Tiger Lake's strong single-thread performance and limited TDP. In fact, Intel already has its Xeon W-11000-series CPUs that support features like ECC that are required by servers. In addition, machines like the reServer are based on an 11th-Gen chip and are designed for edge computing applications and can be classified as servers.

In any case, since Intel has submitted CPU IDs for its Tiger Lake Server products to the GPU-Z developers, it looks like the company is indeed preparing new SKUs that will be available.

TechPowerUp has the full list of GPU-Z v2.42.0 improvements, which includes things like added support for Nvidia's GeForce RTX 3050 Ti Mobile (GA106), T1200 Mobile, GRID K340, GRID M30, Q12U-1; AMD's Radeon Pro W6800X, more stable readings from EVGA's iCX sensor; fixed memory clock readings on some AMD APUs; and so on.